Introduction to the voteogram package

Getting Vote Data

The cartograms need data and the best way to do that is by obtaining roll call vote data from ProPublica via the roll_call() function. Data can be retrieved for any House or Senate vote by specificing the target vote parameters:

sen <- roll_call("senate", 115, 1, 110)
rep <- roll_call("house", 115, 1, 256)

Their structures look the same and there is a print-method to make the console output easier on the eyes:

sen
#> 115th Congress / Session: 1 / Senate Roll Call: 110 / April  6, 2017
#> 
#> Neil M. Gorsuch, of Colorado, to be an Associate Justice of the Supreme Court of the United States
#> 
#> Result: Cloture Motion Agreed to

sen$votes
#> # A tibble: 100 x 11
#>    bioguide_id role_id        member_name  sort_name party state_abbrev
#>  *       <chr>   <int>              <chr>      <chr> <chr>        <chr>
#>  1     A000360     526   Lamar  Alexander  Alexander     R           TN
#>  2     B001230     481      Tammy Baldwin    Baldwin     D           WI
#>  3     B001261     498      John Barrasso   Barrasso     R           WY
#>  4     B001267     561     Michael Bennet     Bennet     D           CO
#>  5     B001277     535 Richard Blumenthal Blumenthal     D           CT
#>  6     B000575     547         Roy  Blunt      Blunt     R           MO
#>  7     B001288     507       Cory  Booker     Booker     D           NJ
#>  8     B001236     551      John  Boozman    Boozman     R           AR
#>  9     B000944     480     Sherrod  Brown      Brown     D           OH
#> 10     B001135     555    Richard M. Burr       Burr     R           NC
#> # ... with 90 more rows, and 5 more variables: display_state_abbrev <chr>,
#> #   district <chr>, position <chr>, dw_nominate <lgl>, pp_id <chr>

rep
#> 115th Congress / Session: 1 / House Roll Call: 256 / May  4, 2017
#> 
#> American Health Care Act
#> 
#> Result: Passed

fortify(rep)
#> # A tibble: 435 x 11
#>    bioguide_id role_id        member_name sort_name party state_abbrev
#>  *       <chr>   <int>              <chr>     <chr> <chr>        <chr>
#>  1     A000374     274      Ralph Abraham   Abraham     R           LA
#>  2     A000370     294        Alma  Adams     Adams     D           NC
#>  3     A000055     224 Robert B. Aderholt  Aderholt     R           AL
#>  4     A000371     427       Pete Aguilar   Aguilar     D           CA
#>  5     A000372     268         Rick Allen     Allen     R           GA
#>  6     A000367     131       Justin Amash     Amash     R           MI
#>  7     A000369     388        Mark Amodei    Amodei     R           NV
#>  8     A000375     320    Jodey Arrington Arrington     R           TX
#>  9     B001291     590        Brian Babin     Babin     R           TX
#> 10     B001298     206          Don Bacon     Bacon     R           NE
#> # ... with 425 more rows, and 5 more variables:
#> #   display_state_abbrev <chr>, district <int>, position <chr>,
#> #   dw_nominate <lgl>, pp_id <chr>

That data may be useful on its own (ouside of plotting).

Note, also, that ggplot2’s fortify() method uses the provided object class method for roll call objects to know how to extract the rectangular data necessary for plotting.

Making Cartograms

These cartograms have a few style options:

senate_carto(sen) +
  labs(title="Senate Vote 110 - Invokes Cloture on Neil Gorsuch Nomination") +
  theme_voteogram()

house_carto(rep, pp_square=TRUE) +
  labs(x=NULL, y=NULL, 
       title="House Vote 256 - Passes American Health Care Act,\nRepealing Obamacare") +
  theme_voteogram()

house_carto(rep, pp_square=FALSE) +
  labs(x=NULL, y=NULL, 
       title="House Vote 256 - Passes American Health Care Act,\nRepealing Obamacare") +
  theme_voteogram()

house_carto(rep, "gt") +
  labs(x=NULL, y=NULL, 
       title="House Vote 256 - Passes American Health Care Act,\nRepealing Obamacare") +
  theme_voteogram()

senate_carto(sen) + theme_voteogram(legend=FALSE)

house_carto(rep) + theme_voteogram(legend=FALSE)

house_carto(rep, pp_square=TRUE) + theme_voteogram(legend=FALSE)